EVIDENCE November, 2011 from to POLICY a note series on learning what works, from the Human Development Network Do vouchers for job training programs help? 65766 Youth unemployment is a problem in many developing velopment Goal to achieve productive employment for countries, where labor market opportunities may be fur- everyone and cut poverty and hunger. As part of this, we ther squeezed by rapid rural migration into big cities, the support research into the effectiveness of jobs programs as cost of higher education and job seekers’ limited informa- a way to help policymakers and governments devise the tion about the best opportunities. Policymakers and aid best for their target populations. In Kenya, the World Bank organizations trying to reduce youth unemployment have supported a special voucher program for young adults to a variety of approaches they can use, including vocational evaluate demand for vocational training and the impact of training programs to give job seekers necessary skills to be training on job seekers. The evaluation showed that offer- employed. But what works best is still not clear: Should ing young adults vouchers that cover program costs does governments subsidize job training programs for young encourage young adults to enroll, and that those who can adults? Should they limit subsidies to public sector pro- use the voucher for a private training program are more grams or include the private sector? Do government-run likely to sign up and stay in school. The results of the study and private institutions differ in their ability to attract stu- are in line with research done in Latin America,1 indicat- dents and prepare them to find a job? What roles do out- ing the benefits of making vocational training more widely side factors, such as distance to a school, play in a young available in developing countries. While further research adult’s decision to enroll in a training program? of this program is planned in Kenya, the initial evaluation At the World Bank, we are actively working to help does show that vouchers for vocational schools – especially countries create and implement viable employment pro- when private sector schools are included – can be an effec- grams, in line with the United Nations Millennium De- tive way to give job seekers employable skills. Case Study Kenya The Technical and Vocational Vouchers Program was young adults were invited to an informational session. A to- launched in 2008 to study the effect of vouchers on participa- tal of 2,705 individuals attended one of the 70 preliminary tion in vocational training programs and the short-term im- meetings, during which they were surveyed on their views of pact on job seekers’ employment choices, jobs and incomes. projected future earnings with and without vocational train- Drawing on a separate panel survey the authors have been ing. They were asked to return two weeks later with a letter collecting in mainly western Kenya, 10,767 out-of-school of support from a local authority and list of preferred schools and courses, a process that helped ensure program partici- Did You Know… pants had a genuine interest in vocational training. • 60% of the Kenyan population is under the age of 30 The final applicant group included 2,163 young adults, • The Kenyan unemployment rate is approximately 40% aged 18 to 30. Half were randomly selected to receive • An estimated 64% of unemployed Kenyans are youth a voucher and the other half placed in a control group. The Youth Banner (Youth advocacy organization), Nairobi, Kenya, Among those picked to receive a voucher, half were given a http://www.youthbanner.org/ voucher that could be used only in a public institution, and 1 Bettinger Eric, Michael Kremer and Juan Saaveda (2010), “Are Educational Vouchers Only Redistributive?� Economic Journal, 20 (546), F204-F228. half received a voucher that could be used in either a public treatment and control groups on labor market expectations, or private vocational training institution. In both cases, the outcomes, course completion and course satisfaction. voucher was worth about $460, which covered (or nearly To evaluate whether information about wage rates and covered) the cost of a public or private vocational program. employment options played a role in a person’s vocational Among program participants, the average years of schooling training choices, researchers included an informational prior to the program was 8.8 years, with 27 percent having component to the study. Half of participants in the voucher dropped out before 8th grade and 24 percent having com- group and half in the control group were told, for example, pleted secondary school. Sixty-three percent of participants which jobs had the highest wages. Women were also shown were female. videos of female students in traditionally male-dominated Researchers collected information on participating train- trades, such as car repair, and were given information about ing institutions, including hours of instruction, facilities, the potentially higher wages in those fields. job placement services and internship availability. Unan- Further research is planned on the impact that vocation- nounced visits were carried out to monitor student progress al training has on medium-to-long term labor market re- and retention. Information on the specifics of training center turns. In addition, research will include analyzing the expe- programs was gathered from the school administrations. In riences of participants in the labor market based on baseline the short-term follow-up survey, information was collected characteristics, and will compare more broadly the relative from a random sample of more than 300 individuals in the effectiveness of public and private institutions. The Findings* The vouchers were very successful at getting But participants who received an unrestricted young adults to enroll in vocational training voucher—one that could be used for either a programs. public or private training program—were more likely to enroll and less likely to drop-out of a Seventy-four percent of participants who received vouch- program than those who received the restrict- ers enrolled in some type of vocational training, compared ed (public institution-only) voucher. with less than four percent of those in the control group. Among those who received vouchers, there was no Seventy-nine percent of people who received the unre- statistically significant difference in use based on gender, stricted voucher attended a vocational training program, age group, years since last schooling or previous vocational compared with 69 percent of those who received a voucher training. There was, however, a difference in use among good only for government-run institutions. those who reported being employed at the time the pro- gram started—they were 4.5 percentage points less likely The 10 percentage point enrollment gap may be to use the voucher. explained partly by the greater choice of schools available to holders of an unrestricted voucher. This bulletin summarizes the results of the research paper “Vocational Education Voucher Delivery and Labor Market Returns: A Randomized Evalu- ation Among Kenyan Youth,� by Joan Hamory Hicks, Michael Kremer, Isaac Mbiti and Edward Miguel. Phase I of the impact evaluation was financed through the Bank-Netherlands Partnership Program (BNPP), and Phase II through the Spanish Impact Evaluation Trust Fund (SIEF). The evaluation was undertaken during the period 2008-2011. For more information contact: Shobhana Sosale, AFTED, ssosale@worldbank.org. 2 Duflo, Esther (2001). “Schooling and Labor Market Consequences of School Construction in Indonesia: Evidence from an Unusual policy Experiment.� American Economic Review, 91(4), 795-813 However, regardless of the type of voucher, participants who had not completed secondary school were less likely to drop out than those who had. This result indicates that vocational training may be an alternative or second-choice path for helping promote better employment among young adults with less school- ing. The lower drop-out rate from vocational programs may indicate that youth with less education see more to gain from sticking with a job training program than those with higher levels of education. Other research2 has shown that cutting the distance to Voucher winners were not deterred from enroll- schools by building more facilities can boost enrollment. ing by information that contradicted their often In the Kenya project, individuals who received an unre- inflated views about the wages they could stricted voucher generally had about six training institu- receive after vocational training. tions to choose from within a 10-kilometer radius, com- pared with two institutions for individuals who could only Participants were overly optimistic in their projection of use their voucher for public institutions. Overall, youth the effect of vocational training on income. On average, with vouchers were at least 12 percentage points more like- they believed that average returns (in the form of earn- ly to enroll in a training institution if there was a private ings) from training were 61 percent; in reality, the return institution within 3 kilometers of them, with suggestive would be around 37 percent. Being told this did not af- evidence that this was driven by those with unrestricted fect people’s interest in applying to the program, nor did vouchers. Together, this suggests that the greater choice of it affect enrollment among voucher winners. nearby schools might partly explain the better take-up rate by those with unrestricted vouchers. Students who received vouchers that could only be used for public training programs also were 16 percentage points more likely to drop out than those who had an unrestricted voucher. One reason could be that the more limited choice for those with the public-only voucher made it more dif- ficult for the students to find (nearby) programs that suited their individual needs in terms of programs or in- stitutional quality. On the other hand, participants with Voucher users’ vocational training choices, by popularity: unrestricted vouchers had a greater variety of schools and courses to choose from, which may have promot- 37 %—Tailoring 18 %—Mechanic ed better matches between students and schools—and 9 %—Hairdressing could therefore be more likely to promote better out- 7 %—Driving comes such as retention. However, there are no mea- 6 %—Masonry sures of learning yet. But women shown videos of women working in In terms of the short-run impact on participants’ traditionally-male jobs, such as auto repair, and ability to find work, anecdotal evidence indicates told that wages were higher in such fields, were that those in a training program were able to more likely to use their vouchers differently. find part-time work because of their enrollment. Given the information women were almost nine percentage Most participants were still in school at the time of this points more likely to express interest in a male-dominated survey, but anecdotal evidence collected during on-site vis- course (especially younger and more educated women) and its to the schools indicated that students seemed to be re- 5 percentage points more likely to enroll. cruited for part-time jobs in their fields of study. The next round of research will look at the medium and long-term labor market returns of vocational training. Conclusion Making policy from evidence High rates of youth unemployment are a social and an tution, which suggests that giving people choice allows them economic problem in many developing countries and job to pick the programs that best meet their needs, be it in terms training programs are one proposed method for boosting of the course itself or distance from their home. While the employment. But getting youth to enroll is not always easy: medium and long-term impact of the program still needs to there can be opportunity costs in terms of lost wages during be studied—most voucher recipients were still enrolled in a the period of schooling, plus there is the cost of the pro- training program when this research was carried out—it is gram itself. These two factors, coupled with what can be the clear that voucher programs can boost demand for vocational difficulties of finding transportation to a far-away school or training and that youth are more likely to enroll and stay in a childcare responsibilities, can work against enrollment. program when their choice of institution is unrestricted. This The Kenyan voucher program shows that youth will take underscores the potential usefulness of including the private advantage of job training when the costs are covered through sector in public policy programs designed to boost demand vouchers. And among those who do enroll, a majority stay for job training. Those working on future programs also may with the program. But retention is boosted among those who want to keep in mind that integrating childcare, or cutting are able to choose between a public or private vocational insti- costs further, could help with program take-up. The Human Development Network, part of the World Bank Group, supports and disseminates research evaluating the impact of development projects to help alleviate poverty. The goal is to collect and build empirical evidence that can help govern- ments and development organizations design and implement the most appropriate and effective policies for better educational, health and job opportunities for people in developing countries. For more information about who we are and what we do, go to: http://www.worldbank.org/hdchiefeconomist The Evidence to Policy note series is produced with the generous support of the Spanish Impact Evaluation Fund (SIEF) THE WORLD BANK, HUMAN DEVELOPMENT NETWORK 1818 H STREET, NW WASHINGTON, DC 20433 Produced by Office of the Chief Economist, Human Development Network, Communications/Aliza Marcus